------------------------------------------------------------------------------------------------------ log: c:\Imbook\bwebpage\Section6\mma25p3extra.txt log type: text opened on: 26 May 2005, 11:33:04 . . ********** OVERVIEW OF MMA25P3EXTRA.DO ********** . . * STATA Program . * copyright C 2005 by A. Colin Cameron and Pravin K. Trivedi . * used for "Microeconometrics: Methods and Applications" . * by A. Colin Cameron and Pravin K. Trivedi (2005) . * Cambridge University Press . . * Chapter 25.8 pages 889-893 . * Evaluating treatment effect of training on Earnings . * This program provides additional analysis and data not in the book . * (1) Compare NSW experiment treated to NSW experiment controls . * (2) Compare NSW experiment treated to CPS "controls" . * [Same as text except "controls" are from CPS not PSID] . . * The program is based on . * MMA25P2MATCHING.DO propensity score matching . . * To run this program you need STATA data files . * nswre74_treated.dta NSW Treated sample . * nswre74_control.dta NSW Control sample (not analyzed earlier) . * propensity_cps.dta CPS Control sample (rather than PSID) . . * To run this program you need the Stata add-ons . * pscore.ado, atts.ado, attr.ado, attnd.ado, attnw.ado . * due to Sascha O. Becker and Andrea Ichino (2002) . * "Estimation of average treatment effects based on propensity scores", . * The Stata Journal, Vol.2, No.4, pp. 358-377. . . * This program uses version 2.02 May 13 2005 for Stata version 8 . * downloadable from http://www.iue.it/Personal/Ichino/#pscore . * We earlier used version 1.29 October 8 2002 for Stata version 7 . * downloadable from http://www.iue.it/Personal/Ichino/#pscore . * and obtained the same results . . * To speed up the program reduce breps: the number of bootstrap . * replications used to obtain bootstrap standard errors . * Bootstrap se's will differ from text as here seed is set to 10101 . . ********** STATA SETUP ********** . . set more off . version 8 . set scheme s1mono /* Used for graphs */ . . ********** DATA DESCRIPTION ********** . . * Data originally from DW99 . * R.H. Dehejia and S. Wahba (1999) . * "Causal Effects in Nonexperimental Studies: reevaluating the . * Evaluation of Training Programs", JASA, 1053-1062 . * or DW02 . * R.H. Dehejia and S. Wahba (2002) . * "Propensity-score Matching Methods for Nonexperimental Causal . * Studies", ReStat, 151-161 . * which in turn are from . * Lalonde, R. (1986), "Evaluating the Econometric Evaluations of . * Training Programs with Experimental Data," AER, 604-620. . . * nswre74_treated.dta N=185 NSW Treated sample only . * nswre74_control.dta N=260 NSW Control sample only . * propensity_cps.dta N=16177 NSW Treated + CPS Control sample (Full CPS or CPS-1) . . ********** (1) ANALYSIS: NSW TREATED VERSUS NSW CONTROLS ********** . . * Read in NSW treated and control and combine . use nswre74_treated.dta, clear . append using nswre74_control.dta . . ** Summarize these data . sum Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- treat | 445 .4157303 .4934022 0 1 age | 445 25.37079 7.100282 17 55 edu | 445 10.19551 1.792119 3 16 black | 445 .8337079 .3727617 0 1 hisp | 445 .0876404 .2830895 0 1 -------------+-------------------------------------------------------- married | 445 .1685393 .3747658 0 1 nodegree | 445 .7820225 .4133367 0 1 re74 | 445 2102.265 5363.582 0 39570.68 re75 | 445 1377.138 3150.961 0 25142.24 re78 | 445 5300.764 6631.492 0 60307.93 -------------+-------------------------------------------------------- u74 | 445 .2674157 .4431092 0 1 u75 | 445 .3505618 .4776829 0 1 . bysort treat: sum ---------------------------------------------------------------------------------------------------- -> treat = 0 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- treat | 260 0 0 0 0 age | 260 25.05385 7.057745 17 55 edu | 260 10.08846 1.614325 3 14 black | 260 .8269231 .3790434 0 1 hisp | 260 .1076923 .3105893 0 1 -------------+-------------------------------------------------------- married | 260 .1538462 .3614971 0 1 nodegree | 260 .8346154 .3722439 0 1 re74 | 260 2107.027 5687.906 0 39570.68 re75 | 260 1266.909 3102.982 0 23031.98 re78 | 260 4554.801 5483.836 0 39483.53 -------------+-------------------------------------------------------- u74 | 260 .25 .4338478 0 1 u75 | 260 .3153846 .4655651 0 1 ---------------------------------------------------------------------------------------------------- -> treat = 1 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- treat | 185 1 0 1 1 age | 185 25.81622 7.155019 17 48 edu | 185 10.34595 2.01065 4 16 black | 185 .8432432 .3645579 0 1 hisp | 185 .0594595 .2371244 0 1 -------------+-------------------------------------------------------- married | 185 .1891892 .3927217 0 1 nodegree | 185 .7081081 .4558666 0 1 re74 | 185 2095.574 4886.62 0 35040.07 re75 | 185 1532.055 3219.251 0 25142.24 re78 | 185 6349.144 7867.402 0 60307.93 -------------+-------------------------------------------------------- u74 | 185 .2918919 .4558666 0 1 u75 | 185 .4 .4912274 0 1 . . * Write data to a text (ascii) file so can use with programs other than Stata . outfile treat age edu black hisp married nodegree re74 re75 re78 u74 u75 /* > */using nswre74_all.asc, replace . . ** Calculate the benchmark Treatment Effect . ** Same as DW02 Tables 2 and 3 NSW row second last column . ** and is the number given in CT page 894 second last line . . regress re78 treat Source | SS df MS Number of obs = 445 -------------+------------------------------ F( 1, 443) = 8.04 Model | 348013183 1 348013183 Prob > F = 0.0048 Residual | 1.9178e+10 443 43290369.3 R-squared = 0.0178 -------------+------------------------------ Adj R-squared = 0.0156 Total | 1.9526e+10 444 43976681.9 Root MSE = 6579.5 ------------------------------------------------------------------------------ re78 | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- treat | 1794.342 632.8534 2.84 0.005 550.5745 3038.11 _cons | 4554.801 408.0459 11.16 0.000 3752.855 5356.747 ------------------------------------------------------------------------------ . . ********** (2) ANALYSIS: NSW TREATED VERSUS CPS CONTROLS ********** . . * This data set has NSW treated and full CPS controls . use propensity_cps.dta, clear . . * Variables u74, u75 were evaluated wrongly in the original file . * So make the following correction . drop u74 u75 . gen u74=0 . replace u74=1 if re74==0 (2044 real changes made) . gen u75=0 . replace u75=1 if re75==0 (1859 real changes made) . gen age2=age*age . gen age3=age2*age . gen edu2=edu*edu . gen edure74=edu*re74 . * Not sure whether this is needed . * Does DW99 use edu*re74*age3 or separately edu*re74 and age3 ? . gen edre74age3=edu*re74*age3 . . ** Summarize these data . sum Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- treat | 16177 .011436 .1063292 0 1 age | 16177 33.14051 11.03651 16 55 edu | 16177 12.00828 2.868005 0 18 black | 16177 .0823391 .2748892 0 1 hisp | 16177 .0718922 .2583173 0 1 -------------+-------------------------------------------------------- married | 16177 .7057551 .4557167 0 1 nodegree | 16177 .3005502 .4585115 0 1 re74 | 16177 13880.47 9613.115 0 35040.07 re75 | 16177 13512.21 9313.207 0 25243.55 re78 | 16177 14749.48 9670.996 0 60307.93 -------------+-------------------------------------------------------- u74 | 16177 .1263522 .3322562 0 1 u75 | 16177 .1149162 .3189307 0 1 age2 | 16177 1220.09 783.4604 256 3025 age3 | 16177 48988.49 45032.59 4096 166375 edu2 | 16177 152.4238 67.06033 0 324 -------------+-------------------------------------------------------- edure74 | 16177 169452.3 129585.8 0 490561 edre74age3 | 16177 9.53e+09 1.21e+10 0 7.75e+10 . bysort treat: sum ---------------------------------------------------------------------------------------------------- -> treat = 0 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- treat | 15992 0 0 0 0 age | 15992 33.22524 11.04522 16 55 edu | 15992 12.02751 2.870846 0 18 black | 15992 .0735368 .2610237 0 1 hisp | 15992 .072036 .2585556 0 1 -------------+-------------------------------------------------------- married | 15992 .7117309 .4529712 0 1 nodegree | 15992 .2958354 .4564316 0 1 re74 | 15992 14016.8 9569.796 0 25862.32 re75 | 15992 13650.8 9270.403 0 25243.55 re78 | 15992 14846.66 9647.392 0 25564.67 -------------+-------------------------------------------------------- u74 | 15992 .1196223 .3245295 0 1 u75 | 15992 .1093047 .3120308 0 1 age2 | 15992 1225.906 784.7382 256 3025 age3 | 15992 49305.85 45139.01 4096 166375 edu2 | 15992 152.9023 67.16633 0 324 -------------+-------------------------------------------------------- edure74 | 15992 171147.6 129218.8 0 465521.8 edre74age3 | 15992 9.64e+09 1.21e+10 0 7.75e+10 ---------------------------------------------------------------------------------------------------- -> treat = 1 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- treat | 185 1 0 1 1 age | 185 25.81622 7.155019 17 48 edu | 185 10.34595 2.01065 4 16 black | 185 .8432432 .3645579 0 1 hisp | 185 .0594595 .2371244 0 1 -------------+-------------------------------------------------------- married | 185 .1891892 .3927217 0 1 nodegree | 185 .7081081 .4558666 0 1 re74 | 185 2095.574 4886.62 0 35040.07 re75 | 185 1532.055 3219.251 0 25142.24 re78 | 185 6349.144 7867.402 0 60307.93 -------------+-------------------------------------------------------- u74 | 185 .7081081 .4558666 0 1 u75 | 185 .6 .4912274 0 1 age2 | 185 717.3946 431.2517 289 2304 age3 | 185 21554.66 20964.71 4913 110592 edu2 | 185 111.0595 39.30388 16 256 -------------+-------------------------------------------------------- edure74 | 185 22898.73 57393.97 0 490561 edre74age3 | 185 4.28e+08 1.24e+09 0 8.75e+09 . . * Write data to a text (ascii) file so can use with programs other than Stata . * This has data as original except for recode of u74 and u75 . outfile treat age edu black hisp married nodegree re74 re75 re78 u74 u75 /* > */ using propensity_cps.asc, replace . . ** Number of replications to use in the bootstrap . ** Ideally at least 400 . global breps 200 . . *** (2A) CPS propensity score model from DW02 Table 2 footnote A . . global CPSDW02 age age2 age3 edu edu2 married nodegree black hisp re74 re75 u74 u75 edure74 . . * With common support option . pscore treat $CPSDW02, pscore(myscore) blockid(myblock) comsup numblo(5) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is treat treat | Freq. Percent Cum. ------------+----------------------------------- 0 | 15,992 98.86 98.86 1 | 185 1.14 100.00 ------------+----------------------------------- Total | 16,177 100.00 Estimation of the propensity score Iteration 0: log likelihood = -1011.0713 Iteration 1: log likelihood = -612.55814 Iteration 2: log likelihood = -481.71035 Iteration 3: log likelihood = -428.3351 Iteration 4: log likelihood = -409.00437 Iteration 5: log likelihood = -404.57736 Iteration 6: log likelihood = -404.16676 Iteration 7: log likelihood = -404.15991 Iteration 8: log likelihood = -404.15991 Logit estimates Number of obs = 16177 LR chi2(14) = 1213.82 Prob > chi2 = 0.0000 Log likelihood = -404.15991 Pseudo R2 = 0.6003 ------------------------------------------------------------------------------ treat | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | 2.425229 .3500652 6.93 0.000 1.739114 3.111344 age2 | -.0672395 .0111308 -6.04 0.000 -.0890555 -.0454234 age3 | .0005685 .0001113 5.11 0.000 .0003505 .0007866 edu | .9247848 .2500694 3.70 0.000 .4346577 1.414912 edu2 | -.0572021 .0136202 -4.20 0.000 -.0838972 -.0305071 married | -1.556471 .2517687 -6.18 0.000 -2.049929 -1.063014 nodegree | .9270591 .3254621 2.85 0.004 .2891651 1.564953 black | 3.850668 .2662868 14.46 0.000 3.328755 4.37258 hisp | 1.673885 .409913 4.08 0.000 .8704705 2.4773 re74 | -.0002203 .0001086 -2.03 0.043 -.0004332 -7.40e-06 re75 | -.0001969 .0000378 -5.21 0.000 -.000271 -.0001228 u74 | 1.749522 .2897311 6.04 0.000 1.18166 2.317385 u75 | .00944 .257531 0.04 0.971 -.4953115 .5141915 edure74 | .0000222 9.08e-06 2.45 0.014 4.43e-06 .00004 _cons | -35.22098 3.797922 -9.27 0.000 -42.66477 -27.77719 ------------------------------------------------------------------------------ note: 3 failures and 0 successes completely determined. Note: the common support option has been selected The region of common support is [.00106139, .93845543] Description of the estimated propensity score in region of common support Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% .0010892 .0010614 5% .001221 .0010615 10% .0013925 .0010625 Obs 4041 25% .0021398 .0010632 Sum of Wgt. 4041 50% .0053823 Mean .0452964 Largest Std. Dev. .1326324 75% .0156111 .9356451 90% .0856723 .93718 Variance .0175914 95% .282253 .9374608 Skewness 4.475994 99% .822637 .9384554 Kurtosis 24.36564 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 8 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** The balancing property is satisfied This table shows the inferior bound, the number of treated and the number of controls for each block Inferior | of block | treat of pscore | 0 1 | Total -----------+----------------------+---------- .0010614 | 3,214 18 | 3,232 .025 | 240 8 | 248 .05 | 172 14 | 186 .1 | 96 19 | 115 .2 | 86 32 | 118 .4 | 31 38 | 69 .6 | 9 20 | 29 .8 | 8 36 | 44 -----------+----------------------+---------- Total | 3,856 185 | 4,041 Note: the common support option has been selected ******************************************* End of the algorithm to estimate the pscore ******************************************* . . * Without common support option . drop myscore myblock . pscore treat $CPSDW02, pscore(myscore) blockid(myblock) numblo(5) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is treat treat | Freq. Percent Cum. ------------+----------------------------------- 0 | 15,992 98.86 98.86 1 | 185 1.14 100.00 ------------+----------------------------------- Total | 16,177 100.00 Estimation of the propensity score Iteration 0: log likelihood = -1011.0713 Iteration 1: log likelihood = -612.55814 Iteration 2: log likelihood = -481.71035 Iteration 3: log likelihood = -428.3351 Iteration 4: log likelihood = -409.00437 Iteration 5: log likelihood = -404.57736 Iteration 6: log likelihood = -404.16676 Iteration 7: log likelihood = -404.15991 Iteration 8: log likelihood = -404.15991 Logit estimates Number of obs = 16177 LR chi2(14) = 1213.82 Prob > chi2 = 0.0000 Log likelihood = -404.15991 Pseudo R2 = 0.6003 ------------------------------------------------------------------------------ treat | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | 2.425229 .3500652 6.93 0.000 1.739114 3.111344 age2 | -.0672395 .0111308 -6.04 0.000 -.0890555 -.0454234 age3 | .0005685 .0001113 5.11 0.000 .0003505 .0007866 edu | .9247848 .2500694 3.70 0.000 .4346577 1.414912 edu2 | -.0572021 .0136202 -4.20 0.000 -.0838972 -.0305071 married | -1.556471 .2517687 -6.18 0.000 -2.049929 -1.063014 nodegree | .9270591 .3254621 2.85 0.004 .2891651 1.564953 black | 3.850668 .2662868 14.46 0.000 3.328755 4.37258 hisp | 1.673885 .409913 4.08 0.000 .8704705 2.4773 re74 | -.0002203 .0001086 -2.03 0.043 -.0004332 -7.40e-06 re75 | -.0001969 .0000378 -5.21 0.000 -.000271 -.0001228 u74 | 1.749522 .2897311 6.04 0.000 1.18166 2.317385 u75 | .00944 .257531 0.04 0.971 -.4953115 .5141915 edure74 | .0000222 9.08e-06 2.45 0.014 4.43e-06 .00004 _cons | -35.22098 3.797922 -9.27 0.000 -42.66477 -27.77719 ------------------------------------------------------------------------------ note: 3 failures and 0 successes completely determined. Description of the estimated propensity score Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% 5.92e-07 1.18e-09 5% 1.72e-06 4.07e-09 10% 3.63e-06 4.24e-09 Obs 16177 25% .0000196 1.55e-08 Sum of Wgt. 16177 50% .0001247 Mean .011436 Largest Std. Dev. .0691037 75% .0010579 .9356451 90% .0073933 .93718 Variance .0047753 95% .0250635 .9374608 Skewness 9.281842 99% .3620009 .9384554 Kurtosis 99.39697 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 13 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** The balancing property is satisfied This table shows the inferior bound, the number of treated and the number of controls for each block Inferior | of block | treat of pscore | 0 1 | Total -----------+----------------------+---------- 0 | 11,635 0 | 11,635 .0007813 | 1,056 2 | 1,058 .0015625 | 932 5 | 937 .003125 | 712 2 | 714 .00625 | 709 2 | 711 .0125 | 306 7 | 313 .025 | 240 8 | 248 .05 | 172 14 | 186 .1 | 96 19 | 115 .2 | 86 32 | 118 .4 | 31 38 | 69 .6 | 9 20 | 29 .8 | 8 36 | 44 -----------+----------------------+---------- Total | 15,992 185 | 16,177 ******************************************* End of the algorithm to estimate the pscore ******************************************* . . * Nearest neighbor matching (random version) . attnd re78 treat $CPSDW02, comsup boot reps($breps) dots logit The program is searching the nearest neighbor of each treated unit. This operation may take a while. ATT estimation with Nearest Neighbor Matching method (random draw version) Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 155 730.380 1049.321 0.696 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches Bootstrapping of standard errors command: attnd re78 treat age age2 age3 edu edu2 married nodegree black hisp re74 re75 u74 u75 > edure74 , pscore() logit comsup statistic: attnd = r(attnd) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attnd | 200 730.3805 1280.829 941.0756 -1125.38 2586.141 (N) | 151.7753 3865.059 (P) | -601.5495 1317.795 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with Nearest Neighbor Matching method (random draw version) Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 155 730.380 941.076 0.776 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches . . * Radius matching: Radius=0.0001 . attr re78 treat $CPSDW02, comsup boot reps($breps) dots logit radius(0.0001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 67 1027 -2935.932 888.041 -3.306 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr re78 treat age age2 age3 edu edu2 married nodegree black hisp re74 re75 u74 u75 e > dure74 , pscore() logit comsup radius(.0001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -2935.932 472.0703 1332.096 -5562.767 -309.0973 (N) | -5186.873 438.6902 (P) | -5999.987 -950.2962 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 67 1027 -2935.932 1332.096 -2.204 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . . * Kernel Matching . attk re78 treat $CPSDW02, comsup boot reps($breps) dots logit The program is searching for matches of each treated unit. This operation may take a while. ATT estimation with the Kernel Matching method --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 3856 1267.716 . . --------------------------------------------------------- Note: Analytical standard errors cannot be computed. Use the bootstrap option to get bootstrapped standard errors. Bootstrapping of standard errors command: attk re78 treat age age2 age3 edu edu2 married nodegree black hisp re74 re75 u74 u75 e > dure74 , pscore() logit comsup bwidth(.06) statistic: attk = r(attk) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attk | 200 1267.716 -64.23519 720.5805 -153.2374 2688.669 (N) | -211.0497 2559.206 (P) | -136.5283 2594.417 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Kernel Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 3856 1267.716 720.580 1.759 --------------------------------------------------------- . . * Stratification Matching . atts re78 treat, pscore(myscore) blockid(myblock) comsup boot reps($breps) dots ATT estimation with the Stratification method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 3856 1505.512 734.270 2.050 --------------------------------------------------------- Bootstrapping of standard errors command: atts re78 treat , pscore(myscore) blockid(myblock) comsup statistic: atts = r(atts) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- atts | 200 1505.512 -9.343635 665.1843 193.7979 2817.227 (N) | 251.7493 2958.461 (P) | 252.6815 2985.052 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Stratification method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 3856 1505.512 665.184 2.263 --------------------------------------------------------- . . *** (2B) CPS propensity score model from DW99 Table 2 footnote A . . global CPSDW99 age age2 edu edu2 nodegree married black hisp re74 re75 u74 u75 edure74 age3 . . * With common support option . drop myscore myblock . pscore treat $CPSDW99, pscore(myscore) blockid(myblock) comsup numblo(5) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is treat treat | Freq. Percent Cum. ------------+----------------------------------- 0 | 15,992 98.86 98.86 1 | 185 1.14 100.00 ------------+----------------------------------- Total | 16,177 100.00 Estimation of the propensity score Iteration 0: log likelihood = -1011.0713 Iteration 1: log likelihood = -612.55814 Iteration 2: log likelihood = -481.71035 Iteration 3: log likelihood = -428.3351 Iteration 4: log likelihood = -409.00437 Iteration 5: log likelihood = -404.57736 Iteration 6: log likelihood = -404.16676 Iteration 7: log likelihood = -404.15991 Iteration 8: log likelihood = -404.15991 Logit estimates Number of obs = 16177 LR chi2(14) = 1213.82 Prob > chi2 = 0.0000 Log likelihood = -404.15991 Pseudo R2 = 0.6003 ------------------------------------------------------------------------------ treat | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | 2.425229 .3500652 6.93 0.000 1.739114 3.111344 age2 | -.0672395 .0111308 -6.04 0.000 -.0890555 -.0454234 edu | .9247848 .2500694 3.70 0.000 .4346577 1.414912 edu2 | -.0572021 .0136202 -4.20 0.000 -.0838972 -.0305071 nodegree | .9270591 .3254621 2.85 0.004 .2891651 1.564953 married | -1.556471 .2517687 -6.18 0.000 -2.049929 -1.063014 black | 3.850668 .2662868 14.46 0.000 3.328755 4.37258 hisp | 1.673885 .409913 4.08 0.000 .8704705 2.4773 re74 | -.0002203 .0001086 -2.03 0.043 -.0004332 -7.40e-06 re75 | -.0001969 .0000378 -5.21 0.000 -.000271 -.0001228 u74 | 1.749522 .2897311 6.04 0.000 1.18166 2.317385 u75 | .00944 .257531 0.04 0.971 -.4953115 .5141915 edure74 | .0000222 9.08e-06 2.45 0.014 4.43e-06 .00004 age3 | .0005685 .0001113 5.11 0.000 .0003505 .0007866 _cons | -35.22098 3.797922 -9.27 0.000 -42.66477 -27.77719 ------------------------------------------------------------------------------ note: 3 failures and 0 successes completely determined. Note: the common support option has been selected The region of common support is [.00106139, .93845543] Description of the estimated propensity score in region of common support Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% .0010892 .0010614 5% .001221 .0010615 10% .0013925 .0010625 Obs 4041 25% .0021398 .0010632 Sum of Wgt. 4041 50% .0053823 Mean .0452964 Largest Std. Dev. .1326324 75% .0156111 .9356451 90% .0856723 .93718 Variance .0175914 95% .282253 .9374608 Skewness 4.475994 99% .822637 .9384554 Kurtosis 24.36564 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 8 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** The balancing property is satisfied This table shows the inferior bound, the number of treated and the number of controls for each block Inferior | of block | treat of pscore | 0 1 | Total -----------+----------------------+---------- .0010614 | 3,214 18 | 3,232 .025 | 240 8 | 248 .05 | 172 14 | 186 .1 | 96 19 | 115 .2 | 86 32 | 118 .4 | 31 38 | 69 .6 | 9 20 | 29 .8 | 8 36 | 44 -----------+----------------------+---------- Total | 3,856 185 | 4,041 Note: the common support option has been selected ******************************************* End of the algorithm to estimate the pscore ******************************************* . . * Without common support option . drop myscore myblock . pscore treat $CPSDW99, pscore(myscore) blockid(myblock) numblo(5) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is treat treat | Freq. Percent Cum. ------------+----------------------------------- 0 | 15,992 98.86 98.86 1 | 185 1.14 100.00 ------------+----------------------------------- Total | 16,177 100.00 Estimation of the propensity score Iteration 0: log likelihood = -1011.0713 Iteration 1: log likelihood = -612.55814 Iteration 2: log likelihood = -481.71035 Iteration 3: log likelihood = -428.3351 Iteration 4: log likelihood = -409.00437 Iteration 5: log likelihood = -404.57736 Iteration 6: log likelihood = -404.16676 Iteration 7: log likelihood = -404.15991 Iteration 8: log likelihood = -404.15991 Logit estimates Number of obs = 16177 LR chi2(14) = 1213.82 Prob > chi2 = 0.0000 Log likelihood = -404.15991 Pseudo R2 = 0.6003 ------------------------------------------------------------------------------ treat | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | 2.425229 .3500652 6.93 0.000 1.739114 3.111344 age2 | -.0672395 .0111308 -6.04 0.000 -.0890555 -.0454234 edu | .9247848 .2500694 3.70 0.000 .4346577 1.414912 edu2 | -.0572021 .0136202 -4.20 0.000 -.0838972 -.0305071 nodegree | .9270591 .3254621 2.85 0.004 .2891651 1.564953 married | -1.556471 .2517687 -6.18 0.000 -2.049929 -1.063014 black | 3.850668 .2662868 14.46 0.000 3.328755 4.37258 hisp | 1.673885 .409913 4.08 0.000 .8704705 2.4773 re74 | -.0002203 .0001086 -2.03 0.043 -.0004332 -7.40e-06 re75 | -.0001969 .0000378 -5.21 0.000 -.000271 -.0001228 u74 | 1.749522 .2897311 6.04 0.000 1.18166 2.317385 u75 | .00944 .257531 0.04 0.971 -.4953115 .5141915 edure74 | .0000222 9.08e-06 2.45 0.014 4.43e-06 .00004 age3 | .0005685 .0001113 5.11 0.000 .0003505 .0007866 _cons | -35.22098 3.797922 -9.27 0.000 -42.66477 -27.77719 ------------------------------------------------------------------------------ note: 3 failures and 0 successes completely determined. Description of the estimated propensity score Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% 5.92e-07 1.18e-09 5% 1.72e-06 4.07e-09 10% 3.63e-06 4.24e-09 Obs 16177 25% .0000196 1.55e-08 Sum of Wgt. 16177 50% .0001247 Mean .011436 Largest Std. Dev. .0691037 75% .0010579 .9356451 90% .0073933 .93718 Variance .0047753 95% .0250635 .9374608 Skewness 9.281842 99% .3620009 .9384554 Kurtosis 99.39697 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 13 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** The balancing property is satisfied This table shows the inferior bound, the number of treated and the number of controls for each block Inferior | of block | treat of pscore | 0 1 | Total -----------+----------------------+---------- 0 | 11,635 0 | 11,635 .0007813 | 1,056 2 | 1,058 .0015625 | 932 5 | 937 .003125 | 712 2 | 714 .00625 | 709 2 | 711 .0125 | 306 7 | 313 .025 | 240 8 | 248 .05 | 172 14 | 186 .1 | 96 19 | 115 .2 | 86 32 | 118 .4 | 31 38 | 69 .6 | 9 20 | 29 .8 | 8 36 | 44 -----------+----------------------+---------- Total | 15,992 185 | 16,177 ******************************************* End of the algorithm to estimate the pscore ******************************************* . . * Nearest neighbor matching (random version) . attnd re78 treat $CPSDW99, comsup boot reps($breps) dots logit The program is searching the nearest neighbor of each treated unit. This operation may take a while. ATT estimation with Nearest Neighbor Matching method (random draw version) Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 155 730.380 1049.321 0.696 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches Bootstrapping of standard errors command: attnd re78 treat age age2 edu edu2 nodegree married black hisp re74 re75 u74 u75 edure > 74 age3 , pscore() logit comsup statistic: attnd = r(attnd) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attnd | 200 730.3805 1179.371 964.5437 -1171.658 2632.419 (N) | -9.143144 3738.959 (P) | -638.1188 1625.387 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with Nearest Neighbor Matching method (random draw version) Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 155 730.380 964.544 0.757 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches . . * Radius matching: Radius=0.0001 . attr re78 treat $CPSDW99, comsup boot reps($breps) dots logit radius(0.0001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 67 1027 -2935.932 888.041 -3.306 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr re78 treat age age2 edu edu2 nodegree married black hisp re74 re75 u74 u75 edure7 > 4 age3 , pscore() logit comsup radius(.0001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -2935.932 522.4813 1276.508 -5453.15 -418.7147 (N) | -5239.598 302.9884 (P) | -6023.029 -1232.031 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 67 1027 -2935.932 1276.508 -2.300 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . . * Kernel Matching . attk re78 treat $CPSDW99, comsup boot reps($breps) dots logit The program is searching for matches of each treated unit. This operation may take a while. ATT estimation with the Kernel Matching method --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 3856 1267.716 . . --------------------------------------------------------- Note: Analytical standard errors cannot be computed. Use the bootstrap option to get bootstrapped standard errors. Bootstrapping of standard errors command: attk re78 treat age age2 edu edu2 nodegree married black hisp re74 re75 u74 u75 edure7 > 4 age3 , pscore() logit comsup bwidth(.06) statistic: attk = r(attk) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attk | 200 1267.716 -57.76407 751.2898 -213.7948 2749.227 (N) | -304.83 2488.355 (P) | -314.1009 2459.423 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Kernel Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 3856 1267.716 751.290 1.687 --------------------------------------------------------- . . * Stratification Matching . atts re78 treat, pscore(myscore) blockid(myblock) comsup boot reps($breps) dots ATT estimation with the Stratification method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 3856 1505.512 734.270 2.050 --------------------------------------------------------- Bootstrapping of standard errors command: atts re78 treat , pscore(myscore) blockid(myblock) comsup statistic: atts = r(atts) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- atts | 200 1505.512 61.77066 741.7862 42.7422 2968.282 (N) | 245.6284 2880.622 (P) | 348.125 2849.896 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Stratification method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 3856 1505.512 741.786 2.030 --------------------------------------------------------- . . *** (2C) CPS propensity score model from Becker-Ichino, 2002 (BI02) . . gen re742 = re74*re74 . gen re752 = re75*re75 . gen blacku74 = black*u74 . global CPSBI02 age age2 edu edu2 married black hisp re74 re75 re742 re752 blacku74 . . * With common support option . drop myscore myblock . pscore treat $CPSBI02, pscore(myscore) blockid(myblock) comsup numblo(5) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is treat treat | Freq. Percent Cum. ------------+----------------------------------- 0 | 15,992 98.86 98.86 1 | 185 1.14 100.00 ------------+----------------------------------- Total | 16,177 100.00 Estimation of the propensity score Iteration 0: log likelihood = -1011.0713 Iteration 1: log likelihood = -660.17479 Iteration 2: log likelihood = -533.64831 Iteration 3: log likelihood = -462.67008 Iteration 4: log likelihood = -435.22392 Iteration 5: log likelihood = -427.14921 Iteration 6: log likelihood = -425.78297 Iteration 7: log likelihood = -425.64689 Iteration 8: log likelihood = -425.64309 Logit estimates Number of obs = 16177 LR chi2(12) = 1170.86 Prob > chi2 = 0.0000 Log likelihood = -425.64309 Pseudo R2 = 0.5790 ------------------------------------------------------------------------------ treat | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | .7902073 .0940972 8.40 0.000 .6057803 .9746344 age2 | -.0128161 .0015894 -8.06 0.000 -.0159313 -.0097009 edu | .9953909 .2558663 3.89 0.000 .4939022 1.49688 edu2 | -.0636036 .0131378 -4.84 0.000 -.0893532 -.0378541 married | -1.534639 .2516679 -6.10 0.000 -2.027899 -1.041379 black | 3.340175 .3032312 11.02 0.000 2.745853 3.934497 hisp | 1.636367 .3971529 4.12 0.000 .8579614 2.414772 re74 | -.0001744 .0000626 -2.79 0.005 -.0002971 -.0000517 re75 | -.000168 .0000693 -2.42 0.015 -.0003039 -.0000322 re742 | 8.06e-09 2.61e-09 3.09 0.002 2.95e-09 1.32e-08 re752 | -2.05e-09 3.97e-09 -0.52 0.605 -9.83e-09 5.73e-09 blacku74 | 1.033264 .288037 3.59 0.000 .4687217 1.597806 _cons | -18.16269 1.865757 -9.73 0.000 -21.81951 -14.50588 ------------------------------------------------------------------------------ note: 112 failures and 0 successes completely determined. Note: the common support option has been selected The region of common support is [.00065577, .90386519] Description of the estimated propensity score in region of common support Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% .0006768 .0006558 5% .0007912 .000656 10% .0009583 .0006562 Obs 5354 25% .0016749 .0006566 Sum of Wgt. 5354 50% .0040446 Mean .0343457 Largest Std. Dev. .1120884 75% .0089357 .8905055 90% .0495031 .898552 Variance .0125638 95% .1913766 .9023286 Skewness 4.931471 99% .6773557 .9038652 Kurtosis 29.27201 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 10 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** Variable blacku74 is not balanced in block 3 The balancing property is not satisfied Try a different specification of the propensity score Inferior | of block | treat of pscore | 0 1 | Total -----------+----------------------+---------- 0 | 4,230 13 | 4,243 .0125 | 330 7 | 337 .025 | 231 9 | 240 .05 | 126 14 | 140 .1 | 108 23 | 131 .2 | 87 30 | 117 .4 | 29 20 | 49 .5 | 10 24 | 34 .6 | 12 25 | 37 .8 | 6 20 | 26 -----------+----------------------+---------- Total | 5,169 185 | 5,354 Note: the common support option has been selected ******************************************* End of the algorithm to estimate the pscore ******************************************* . . * Without common support option . drop myscore myblock . pscore treat $CPSBI02, pscore(myscore) blockid(myblock) numblo(5) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is treat treat | Freq. Percent Cum. ------------+----------------------------------- 0 | 15,992 98.86 98.86 1 | 185 1.14 100.00 ------------+----------------------------------- Total | 16,177 100.00 Estimation of the propensity score Iteration 0: log likelihood = -1011.0713 Iteration 1: log likelihood = -660.17479 Iteration 2: log likelihood = -533.64831 Iteration 3: log likelihood = -462.67008 Iteration 4: log likelihood = -435.22392 Iteration 5: log likelihood = -427.14921 Iteration 6: log likelihood = -425.78297 Iteration 7: log likelihood = -425.64689 Iteration 8: log likelihood = -425.64309 Logit estimates Number of obs = 16177 LR chi2(12) = 1170.86 Prob > chi2 = 0.0000 Log likelihood = -425.64309 Pseudo R2 = 0.5790 ------------------------------------------------------------------------------ treat | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | .7902073 .0940972 8.40 0.000 .6057803 .9746344 age2 | -.0128161 .0015894 -8.06 0.000 -.0159313 -.0097009 edu | .9953909 .2558663 3.89 0.000 .4939022 1.49688 edu2 | -.0636036 .0131378 -4.84 0.000 -.0893532 -.0378541 married | -1.534639 .2516679 -6.10 0.000 -2.027899 -1.041379 black | 3.340175 .3032312 11.02 0.000 2.745853 3.934497 hisp | 1.636367 .3971529 4.12 0.000 .8579614 2.414772 re74 | -.0001744 .0000626 -2.79 0.005 -.0002971 -.0000517 re75 | -.000168 .0000693 -2.42 0.015 -.0003039 -.0000322 re742 | 8.06e-09 2.61e-09 3.09 0.002 2.95e-09 1.32e-08 re752 | -2.05e-09 3.97e-09 -0.52 0.605 -9.83e-09 5.73e-09 blacku74 | 1.033264 .288037 3.59 0.000 .4687217 1.597806 _cons | -18.16269 1.865757 -9.73 0.000 -21.81951 -14.50588 ------------------------------------------------------------------------------ note: 112 failures and 0 successes completely determined. Description of the estimated propensity score Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% 2.89e-08 1.94e-10 5% 3.05e-07 1.94e-10 10% 1.20e-06 1.94e-10 Obs 16177 25% .0000148 1.94e-10 Sum of Wgt. 16177 50% .0001313 Mean .011436 Largest Std. Dev. .0664629 75% .0016513 .8905055 90% .0074369 .898552 Variance .0044173 95% .0234798 .9023286 Skewness 8.811019 99% .3855562 .9038652 Kurtosis 89.82108 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 14 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** Variable blacku74 is not balanced in block 7 The balancing property is not satisfied Try a different specification of the propensity score Inferior | of block | treat of pscore | 0 1 | Total -----------+----------------------+---------- 0 | 11,076 1 | 11,077 .0007813 | 968 2 | 970 .0015625 | 1,020 2 | 1,022 .003125 | 1,185 3 | 1,188 .00625 | 804 5 | 809 .0125 | 330 7 | 337 .025 | 231 9 | 240 .05 | 126 14 | 140 .1 | 108 23 | 131 .2 | 87 30 | 117 .4 | 29 20 | 49 .5 | 10 24 | 34 .6 | 12 25 | 37 .8 | 6 20 | 26 -----------+----------------------+---------- Total | 15,992 185 | 16,177 ******************************************* End of the algorithm to estimate the pscore ******************************************* . . * Nearest neighbor matching (random version) . attnd re78 treat $CPSBI02, comsup boot reps($breps) dots logit The program is searching the nearest neighbor of each treated unit. This operation may take a while. ATT estimation with Nearest Neighbor Matching method (random draw version) Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 147 1214.888 988.298 1.229 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches Bootstrapping of standard errors command: attnd re78 treat age age2 edu edu2 married black hisp re74 re75 re742 re752 blacku74 , > pscore() logit comsup statistic: attnd = r(attnd) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attnd | 200 1214.888 379.5276 924.3417 -607.8733 3037.65 (N) | -199.325 3378.257 (P) | -1646.026 2654.964 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with Nearest Neighbor Matching method (random draw version) Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 147 1214.888 924.342 1.314 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches . . * Radius matching: Radius=0.0001 . attr re78 treat $CPSBI02, comsup boot reps($breps) dots logit radius(0.0001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 65 1089 -3094.104 857.247 -3.609 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr re78 treat age age2 edu edu2 married black hisp re74 re75 re742 re752 blacku74 , > pscore() logit comsup radius(.0001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -3094.104 603.6858 1724.927 -6495.585 307.3775 (N) | -5865.623 247.5659 (P) | -8184.668 -474.5812 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 65 1089 -3094.104 1724.927 -1.794 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . . * Kernel Matching . attk re78 treat $CPSBI02, comsup boot reps($breps) dots logit The program is searching for matches of each treated unit. This operation may take a while. ATT estimation with the Kernel Matching method --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 5169 881.520 . . --------------------------------------------------------- Note: Analytical standard errors cannot be computed. Use the bootstrap option to get bootstrapped standard errors. Bootstrapping of standard errors command: attk re78 treat age age2 edu edu2 married black hisp re74 re75 re742 re752 blacku74 , > pscore() logit comsup bwidth(.06) statistic: attk = r(attk) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attk | 200 881.5195 193.3904 741.3048 -580.3012 2343.34 (N) | -375.8089 2373.732 (P) | -776.3726 2117.355 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Kernel Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 5169 881.520 741.305 1.189 --------------------------------------------------------- . . * Stratification Matching . atts re78 treat, pscore(myscore) blockid(myblock) comsup boot reps($breps) dots ATT estimation with the Stratification method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 5169 1538.713 . . --------------------------------------------------------- Bootstrapping of standard errors command: atts re78 treat , pscore(myscore) blockid(myblock) comsup statistic: atts = r(atts) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 16177 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- atts | 200 1538.713 18.76738 748.4438 62.81438 3014.612 (N) | 249.6562 3263.537 (P) | 225.0108 3230.658 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Stratification method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 5169 1538.713 748.444 2.056 --------------------------------------------------------- . . ********** CLOSE OUTPUT ********** . log close log: c:\Imbook\bwebpage\Section6\mma25p3extra.txt log type: text closed on: 26 May 2005, 13:26:49 ----------------------------------------------------------------------------------------------------